Ultrasound-Based Device to Guide Treatment of Graft- Versus-Host-Disease using Skin Elasticity as a Biomarker 1 ABSTRACT Sclerotic chronic graft versus host disease (cGVHD) develops in 30 - 70% of allogenic Human Stem Cell Transplant (HCT) recipients, and is associated with significant morbidity and mortality. GVHD is treated with immunosuppression, which puts patients at severe risk of infection. Immunosuppression must therefore be delivered judiciously to obtain the therapeutic effect while minimizing dose, but current clinical instruments for quantifying cGVHD provide an incomplete picture and are not sensitive enough to predict outcome or response to therapy[1], [2]. Currently, there is no standard device used to measure skin sclerosis in clinical practice. Indeed, the NIH Chronic GVHD Working Group has stated that: ?There is an urgent need for the development of more quantifiable and reproducible measurements or imaging methods that could be used in patients with sclerotic skin manifestations of chronic GVHD?[3] Our team includes the inventors of ultrasound-based methods for measuring the elasticity of tissue[4], which we have demonstrated to distinguish healthy skin from sclerotic in preliminary clinical studies using our research platform[5]. We have invented a specific type of elasticity measurement called Constructive Shear Wave Interference Velocimetry (CSWIV), optimized for use in portable, point-measurement systems without the complexity and cost of full-scale clinical imaging systems, and with features to improve estimates in highly sclerotic skin over previous methods, and tested it in simulations and a bench-top prototype. Skin is a complex organ, consisting of multiple layers of anisotropic mechanical properties. We propose to use simulations performed using Duke?s FEM and acoustic numeric tools to develop model-based methods for quantifying the absolute mechanical properties of complex, layered media using CSWIV, and use these results to guide acoustic design for a prototype device. Probe alignment was also a challenge in the initial work, so we further propose to develop a transducer alignment-assisting acoustic standoff system. The constructed device will be validated in a multi-user study using tissue-mimicking phantoms. If successful, Phase II will test the device in patients post-HCT to see if the device is sensitive and specific enough to detect the onset of GVHD or the response to therapy in way that could meaningfully impact administration of immunosuppression. 1